Package 'ttdo'

Test for equality with explicit difference

Description

Test for equality with explicit difference

Usage

expect_equal_with_diff(
  current,
  target,
  tol = sqrt(.Machine$double.eps),
  mode = getOption("diffobj.mode", "unified"),
  format = getOption("diffobj.format", "ansi256"),
  ...
)

expect_equivalent_with_diff(
  current,
  target,
  tol = sqrt(.Machine$double.eps),
  ...
)

Arguments

current	[R object or expression] Outcome or expression under scrutiny.
target	[R object or expression] Expected outcome
tol	[numeric] Test equality to machine rounding. Passed to all.equal (tolerance)
mode	[character] Comparison mode passed to diffPrint, defaults to using the “diffobj.mode” global option value with “unified” as fallback if no such option is set
format	[character] Comparison mode passed to diffPrint, defaults to to using the “diffobj.format” global option value with “ansi256” as fallback if no such option is set
...	Passed to all.equal

Details

expect_equivalent_with_diff calls expect_equal_with_diff with the extra arguments check.attributes=FALSE and use.names=FALSE

Value

A tinytest object. A tinytest object is a logical with attributes holding information about the test that was run. The class attribute is set to c("ttdo", "tinytest") to signal that it is a 'diffobj' result.

Examples

library(tinytest)
using(ttdo)
expect_equal_with_diff(1 + 1, 2)		# TRUE
expect_equal_with_diff(1 - 1, 2)		# FALSE
expect_equivalent_with_diff(2, c(x=2))	# TRUE
expect_equivalent_with_diff(2, c(x=2))	# TRUE

---

Extensions of equality tests for tinytest

Description

Building on the tinytest functions for testing equality with optional enhanced object diffing and additional test feedback through addtional attributes.

Usage

expect_equal_xl(
  current,
  target,
  useDiffObj = TRUE,
  tol = sqrt(.Machine$double.eps),
  info = NA_character_,
  mode = getOption("diffobj.mode", "unified"),
  format = getOption("diffobj.format", "ansi256"),
  ...
)

expect_identical_xl(
  current,
  target,
  useDiffObj = TRUE,
  info = NA_character_,
  mode = getOption("diffobj.mode", "unified"),
  format = getOption("diffobj.format", "ansi256"),
  ...
)

expect_equivalent_xl(
  current,
  target,
  useDiffObj = TRUE,
  tol = sqrt(.Machine$double.eps),
  info = NA_character_,
  mode = getOption("diffobj.mode", "unified"),
  format = getOption("diffobj.format", "ansi256"),
  ...
)

Arguments

current	[R object or expression] Outcome or expression under scrutiny.
target	[R object or expression] Expected outcome
useDiffObj	[logical] Whether you should use diffPrint for the diff field in the resulting tinytest object
tol	[numeric] Test equality to machine rounding. Passed to all.equal (tolerance)
info	An additional attribute to pass around with the tinytest object
mode	[character] Comparison mode passed to diffPrint, defaults to using the “diffobj.mode” global option value with “unified” as fallback if no such option is set
format	[character] Comparison mode passed to diffPrint, defaults to to using the “diffobj.format” global option value with “ansi256” as fallback if no such option is set
...	Passed to all.equal and returned as a test attributes

Details

While tinytest does now support the passing of additional information with the info field in its tests, they are not yet supported in the as.data.frame.tinytests method.

Value

A tinytest object. A tinytest object is a logical with attributes holding information about the test that was run

Examples

library(tinytest)
using(ttdo)
expect_equal_xl(1 + 1, 2, score = 3) # TRUE
expect_equal_xl(1 - 1, 2, name = "check 1-1==2", score = 1, totalpts = 2) # FALSE

---

Test for plot equality with 'diff' generation

Description

This tinytest-compatible expectation compares two plots supplied as files (which should be png files) and returns visual difference in a plot file (for which the designated file has to be supplied).

The labels “old” (for the reference plot compared against), “new” (for the candidate plot, and “diff” can be set as value to the global option tinysnapshot_plot_diff_style. For example settings c("old", "new", "diff") shows all three, setting c("new", "diff") just these two. The default is to only show the ‘diff’ plot.

The data format used for returning the PNG image created is described in https://en.wikipedia.org/wiki/Data_URI_scheme.

Usage

expect_visual_equal_with_diff(proposed, reference, difference, ...)

Arguments

proposed	Character value with filename of proposed solution, should be png
reference	Character value with filename of reference plot, should be png
difference	Character value with filename for difference plot (if plots differ)
...	Passes on to tinysnapshot::expect_equivalent_images()

Value

The tinytest result object where the diff attribute contains the suitable value that can be passed onto the JSON output, i.e. a character string beginning with "date:image/png;base64," followed with the base64 encoded file. The class attribute is set to c("ttvd", "tinytest") to signal that it is a 'visual diff' result.

---

Convert tinytest results to data.frame

Description

This method extends the as.data.frame.tinytest method to handle arbitrary attributes attached to each tinytest object. You can pass in the results of a single test (a tinytest object) directly or the results of one of the run_test_* functions (a tinytests object).

Usage

makeDataFrame(x)

Arguments

x	a tinytest or tinytests object

Examples

# create a test file in tempdir
tests <- "
using(ttdo)

addOne <- function(x) x + 2

expect_true(addOne(0) > 0)
expect_equal(2, addOne(1))
"
testfile <- tempfile(pattern = "test_", fileext = ".R")
write(tests, testfile)

# extract testdir
testdir <- dirname(testfile)
# run all files starting with 'test' in testdir
library(tinytest)
out <- run_test_dir(testdir)
#
# convert results
dat <- makeDataFrame(out)
dat

dat2 <- makeDataFrame(expect_equal_xl(1-1, 2, useDiffObj = FALSE, name = 'subtr', pts = 1))

---

Extensions of boolean and messaging tests from tinytest

Description

Building on the tinytest functions for testing boolean values with additional test feedback through attributes.

Usage

expect_true_xl(current, info = NA_character_, ...)

expect_false_xl(current, info = NA_character_, ...)

expect_null_xl(current, info = NA_character_, ...)

expect_silent_xl(current, quiet = TRUE, info = NA_character_, ...)

expect_error_xl(
  current,
  pattern = ".*",
  class = "error",
  info = NA_character_,
  ...
)

expect_warning_xl(
  current,
  pattern = ".*",
  class = "warning",
  info = NA_character_,
  strict = FALSE,
  ...
)

expect_message_xl(
  current,
  pattern = ".*",
  class = "message",
  info = NA_character_,
  strict = FALSE,
  ...
)

Arguments

current	[R object or expression] Outcome or expression under scrutiny.
info	scalar. Optional user-defined message. Must be a single character string. Multiline comments may be separated by "\n".
...	Passed to all.equal and returned as a test attribute
quiet	[logical] suppress output printed by the current expression (see examples)
pattern	[character] A regular expression to match the message.
class	[character] For condition signals (error, warning, message) the class from which the condition should inherit.
strict	[logical] scalar. If set to TRUE, any exception worse than the wanted exception will cause the test to fail.

Details

While tinytest does now support the passing of additional information with the info field in its tests, they are not yet supported in the as.data.frame.tinytests method.

Value

A tinytest object. A tinytest object is a logical with attributes holding information about the test that was run

Examples

library(tinytest)
using(ttdo)
expect_true_xl(TRUE, score = 3) # TRUE
expect_true_xl(FALSE, name = "check 1-1==2", score = 1, totalpts = 2) # FALSE

---